Re-Identification in Video Surveillance Systems Considering Appearance Changes

Abstract

Human behavior analysis and visual anomaly detection are important applications in fields such as video surveillance, security systems, intelligent houses, and elderly care. People re-identification is one of the main steps in a surveillance system that directly affects system performance; and variations in appearance, pose and scene illumination may be challenging issues for such system. Previous re-identification approaches faced limitations while considering appearance changes in their tracking task. This paper proposes a new approach for people re-identification using a descriptor which is robust to appearance changes. In our proposed method the enhanced Gaussian of Gaussian (GOG) and the Hierarchical Gaussian Descriptors (HGDs) are employed to extract feature vectors from images. Experimental results on a number of commonly used people re-identification databases imply the superiority of the proposed approach in people re-identification compared to other existing approaches.